Abstract Many human diseases, including Alzheimer's Disease (AD) are caused by genetic mosaicism, in which a few defective cells carry unique gene mutations distinct from the rest of the body. Therefore, genetic mosaic animal models are highly valuable research tools. For basic research, these models reveal in vivo behaviors of mutant cells to facilitate our understanding of disease etiology; for translational research, they can be used for preclinical testing for drug efficacy in preventing, treating, and even reversing diseases. Previously our lab established a genetic mosaic system in mouse, termed Mosaic Analysis with Double Markers (MADM, Zong 2005 Cell). From a colorless, heterozygous mouse, the MADM system generates sporadic GFP+ mutant and RFP+ WT sibling cells. Sparse labeling and 100% color-genotype matching enable in vivo phenotypic analysis at the single-cell resolution. Mouse MADM was broadly adopted in many fields such as neurobiology (Hippenmeyer 2010 Neuron), developmental biology (Packard 2013 Developmental cell), and cancer biology (Liu 2011 Cell). Compared to mice, zebrafish as a model organism carries great advantages in terms of the transparency of its body and the ease to generate a large, genetically identical population. Therefore, in our funded NIH R21 (OD026524), we proposed to establish a zebrafish equivalent of the mouse MADM system. We have made significant progresses in the past 8 months, including constructing and testing multiple plasmids tailored for the zebrafish MADM system; and successful targeting of the MADM elements into zebrafish genome using Crispr-CAS9 technology. In response to PA-18-591, in this Supplement Application we propose to create additional zebrafish MADM alleles for AD modeling. Specifically, because previously proposed MADM alleles can only be used for gene inactivation, here we propose to add the capacity of gene over-expression into the system that is necessary for AD modeling since it is caused by the mis-expression of mutant APP or Tau. In Specific Aim 1 of this application we propose to create zebrafish MADM lines that allow regulated gene expression at single-cell resolution by slightly modifying the original design. Additionally, considering the prolonged time period of AD progression, in Specific Aim 2 we propose to optimize whole fish, 3D, realtime, long-term imaging at the single-cell resolution of the zebrafish model to enable us to monitor both the progressive defects in mutant protein-expressing neurons and collateral damages to their neighboring glia, immune, and vascular cells. This Supplement Application is within the scope of the original R21 because all proposed experiments are direct extensions of previously designed experiments, which have been progressing smoothly. Moreover, this Supplement Application is highly responsive to PA-18-591 since it will not only enable AD modeling at an unprecedented temporal and spatial resolution for a holistic investigation of disease progression, but also provide a powerful platform for drug screening for the prevention and treatment of AD. Therefore, we expect that our work should lead to a groundbreaking impact on the field of ADRD research.